Learning

Classical and operant conditioning, reinforcement and its schedules, observational, insight and latent learning, verbal learning and the phases of skill learning.

Key ideas

Learning is a relatively permanent change in behaviour produced by experience or practice; performance is what we actually see.
In classical conditioning (Pavlov) a neutral stimulus paired with an unconditioned stimulus comes to produce the response on its own.
In operant conditioning (Skinner) behaviour is shaped by its consequences; the response is instrumental in getting the reward.
Negative reinforcement strengthens a response that removes something unpleasant; punishment weakens a response.
Partial reinforcement makes a response more resistant to extinction than continuous reinforcement.
People also learn by watching models (Bandura), by sudden insight (Köhler) and by latent learning shown only when a reward appears (Tolman).
Skills are learned in three phases: cognitive, associative and autonomous.

Remember

Pavlov	food = US → salivation = UR; bell + food, many times; bell = CS → salivation = CR
CS–US timing	simultaneous, delayed (most effective), trace, backward (rarely works)
Reinforcers	primary: food, water; secondary: money, praise, medals
Rewards on a schedule	every nth response: rewards = responses ÷ n
Key terms	extinction (response fades); spontaneous recovery (returns after rest); generalisation vs discrimination
Verbal learning methods	paired-associates, serial learning, free recall

Watch out

Treating negative reinforcement as punishment: it increases a response, punishment reduces one.
Mixing up the CS and US: the US (food) works without learning; the CS (bell) works only after pairing.
Thinking continuous reinforcement lasts longest once rewards stop; the partial schedule does.

Worked example

Suppose one rat is rewarded for every lever press (continuous) and another for every 5th press (partial). Each presses 100 times. How many rewards does each get, and which rat will keep pressing longer once rewards stop?

Working: 100 × 1; 100 ÷ 5

Answer: 100 and 20 rewards; the rat on the partial schedule (greater resistance to extinction)